The computed tomography imaging spectrometer (“CTIS”) enables spectral imaging of transient events by capturing spatial and spectral information in a single snapshot. That is, the CTIS captures spatial and spectral information from a two-dimensional (“2D”) scene in a single image frame.
In a typical CTIS, as shown in FIGS. 1-3, spots of visible light, namely a blue spot B, a red spot R, and a white spot W, in the field stop 41 are collimated in a lens 32, filtered through a wide-band filter means 33, and passed through a 2D grating disperser 34 which produces a 2D array of diffraction orders 35. A final focusing element, such as a lens 36, re-images the various diffraction orders of light 37 onto a FPA detector 38 (e.g. a charge coupled device (“CCD”)) that records the intensity but not the color of the incident light. Each diffraction order transmitted from the grating disperser 34 produces a spectrally dispersed image 44 of the scene, except for the undiffracted “zeroth” order which produces an undispersed image in the dashed center area 45 of the FPA detector 44, as illustrated best in FIG. 3.
Current systems are generally either slit imaging spectrometers or bandpass-filter imaging spectrometers. However, slit imaging spectrometers must scan the scene spatially to build up a 2D image, and bandpass-filter imaging spectrometers must scan the scene spectrally. The CTIS captures the scene's spatial and spectral information by imaging the scene through a 2D grating disperser, as discussed above and illustrated in FIGS. 1-3. This produces multiple, spectrally dispersed images of the scene that are recorded by a focal plane array (“FPA”) detector. From the captured intensity pattern, computed-tomography algorithms can be used to reconstruct the scene into a cube of spatial (x and y) and spectral (wavelength) information.
The non-scanning nature of the CTIS enables transient-event imaging spectrometry and thus opens up new applications that were previously impossible due to scene movement/evolution during scanning. These include for example: 1) spectral imaging of living biological systems that move/change rapidly during an experiment (e.g. cells, retina, colon, etc.); 2) industrial processes such as semiconductor etching; 3) defense surveillance or regions in which neither the location nor the time of an explosion, missile launch, or chem-bio weapon deployment is known. In addition, the CTIS can be used for static scene spectral imaging when the spatial and spectral resolution requirements are not too demanding.
Current imaging spectrometers use monochrome cameras for capturing the spectrally dispersed images that are used to reconstruct the spatial-spectral information in the scene being imaged. Monochrome camera CTIS systems have scene-dependent spectral resolution and tomographic reconstruction artifacts. This is largely because the reconstruction algorithm does not have enough information to effectively sort out the overlapping information in the spectrally dispersed diffraction images. When a scene does not have significant spatial or spectral diversity, the dispersed images are very smooth, without structural features. This lack of structure causes the reconstruction algorithm to stagnate with a poor solution to the spatial-spectral data cube because a poor solution has nearly the same error as the correct solution. In other words, the reconstruction merit function for these types of scenes has a very broad minimum, so poor solutions are not effectively rejected.